CN102521587A - License plate location method - Google Patents
License plate location method Download PDFInfo
- Publication number
- CN102521587A CN102521587A CN2011103790111A CN201110379011A CN102521587A CN 102521587 A CN102521587 A CN 102521587A CN 2011103790111 A CN2011103790111 A CN 2011103790111A CN 201110379011 A CN201110379011 A CN 201110379011A CN 102521587 A CN102521587 A CN 102521587A
- Authority
- CN
- China
- Prior art keywords
- image
- license plate
- vehicle image
- line segment
- color
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a license plate location method, which comprises the following steps of: (1) preprocessing an original vehicle image I (x,y) to obtain an enhanced vehicle image I1 (x,y); (2) correcting the color values of points in the vehicle image I1 (x,y) to obtain a corrected vehicle image I2 (x,y); (3) scanning the corrected vehicle image I2 (x,y), and extracting a blue-and-yellow-concentrated region as a license plate candidate region; and (4) checking a license plate region, i.e. judging the size of the current region, accounting the current region as the license plate region if the region is in a set range, and finishing license plate location. With the adoption of the license plate location method disclosed by the invention, the license plates in vehicle images acquired under general conditions can be located, and the license plates in vehicle images photographed under complicated conditions, such as dim light, rain, snow, smoke and the like, can also be located, so that the license plate location method has important practical value in license plate recognition.
Description
Technical field
The invention belongs to the computer picture recognition field, be specifically related to a kind of license plate locating method.
Background technology
Car plate identification (license plate recognition; LPR) technology is intelligent transportation system (intelligent transport system; ITS) important means of management, its task are through gathering, analyze, handle image location and identification vehicle license plate automatically.The LPR system can be widely used in important events such as public security bayonet socket, parking lot management, road management violating the regulations, highway inspection, the black board vehicle of monitoring.Car plate location in the Vehicle License Plate Recognition System is that car plate is cut apart, the basis of character recognition, also is the committed step that improves the car plate discrimination.Many scholars are in exploration and the research carried out aspect the car plate location.At present, the method for car plate location mainly be divided into based on gray level image and based on two big types of coloured image.
Based on the license plate locating method of gray level image, the coloured image that before this equipment was collected converts gray level image to, abandons the colouring information of image, take all factors into consideration the information such as gray scale, texture, edge of license plate area again, realizes the location of car plate.This localization method is simple, quick, particularly to clear picture, and the uncomplicated license plate image of background, accuracy rate is higher, but is directed to background complicacy, the weak license plate image of contrast, and this localization method location car plate just compares difficulty.
Given this, scholars forward research emphasis on the license plate locating method based on coloured image that does not abandon color of image information to.This method is the direct coloured image that collects of use equipment, realizes having made full use of the image information that collects in the car plate location; Improved the accuracy rate of car plate location, still since the difference of image capture device with influenced by extraneous factor; Make the picture quality that collects be difficult to accurately tolerance and prediction; And it is bigger influenced by picture quality based on the license plate locating method of coloured image, particularly to strong, the dark partially situation of picture contrast, is difficult to accurately locate car plate.
Theoretical according to Retinex, the color of object is to be determined the reflection potential of long wave, medium wave and shortwave is common by object, and the reflection potential of object in certain wave band is the intrinsic attribute of object itself, do not have dependence with light source.
Summary of the invention
To above problem; The invention provides a kind of license plate locating method based on coloured image; Through considering the light power to the influence of vehicle image and the demand of coloured image license plate locating method, irradiation component and reflecting component ratio in the adjustment vehicle image realize the identification positioning to car plate.
Realize that the concrete numerical procedure that the object of the invention adopted is following:
Step 1, to the original vehicle image I (x y) carries out pre-service, the image I 1 after being enhanced (x, y);
Step 2, correction image I1 (x, y) in the color value of each point, obtain revising back vehicle image I2 (x, y);
(x y), extracts blueness, yellow zone of concentrating as the vehicle candidate region for step 3, scan image I2.
Step 4, verification vehicle region.Judge the current region size, if the zone is then thought license plate area, otherwise cast out within the scope of setting.
The present invention has taken all factors into consideration the influence of the strong and weak vehicle image of light and the demand of coloured image vehicle positioning method; Irradiation component and reflecting component ratio in the vehicle image have been adjusted; Image overall contrast and local contrast have been improved; Strengthen vehicle image shade details; Overcome tradition based on the coloured image license plate locating method can not locate accurately that illumination is poor, the contrast problem of strong image not, enlarged usable range based on coloured image location car plate, promoted the accuracy of car plate location.
Description of drawings
Fig. 1 is overall flow figure of the present invention;
Fig. 2 is image pretreatment process figure;
Fig. 3 uses vehicle image for instance;
Fig. 4 is the vehicle image behind adjustment irradiation component and the reflecting component;
Fig. 5 is the vehicle image after the correction color;
Fig. 6 is the car plate of the inventive method location.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is described in further detail.
Step 1, image pre-service, flow process is as shown in Figure 2, and result is as shown in Figure 4.
(1) vehicle image (as shown in Figure 3) function table is shown as following form:
I(x,y)=R(x,y)L(x,y)
Wherein (x y) is the position of image mid point, and I representes the original vehicle image, and R representes reflected light component, expression irradiates light component, and irradiates light component L describes the brightness of surrounding environment, and is irrelevant with scenery; And reflected light component to be R refer to scenery reflection potential, irrelevant with illumination, it has comprised the detailed information of scenery.
(2) taken the logarithm in the function both sides, so with the irradiates light component and reflected light component is expressed as and form:
LogI=Log(R.L)=LogR+LogL
(3) with Gauss's template of n different scale original image I is done convolution, each convolution is equivalent to original image is done LPF one time, can obtain the image D behind the n width of cloth LPF
1, D
2, D
3D
n,
The expression yardstick is σ
iThe gaussian filtering function, i ∈ [1, n] (preferred n=3 in the present embodiment, σ
1, σ
2, σ
3Preferably be respectively 10,50,240):
(4) in log-domain, the image subtraction with behind original image and the every width of cloth LPF just can obtain the image that the n panel height strengthens frequently:
R
i(x,y)=LogI(x,y)-LogD
i(x,y)
(5) the n panel height that obtains in (4) is strengthened the image weighted sum frequently:
ω
iBe the weight that the i panel height strengthens image frequently, (x y) is vehicle image (ω in this example after the enhancing to I1
1=ω
2=ω
3=1/3).
Step 2, correction image I1 (x, y) in the color value of each point obtain I2 (x, y), revised result be as shown in Figure 5.
(1) image is transformed into the HSV space from rgb space, in the hsv color model in each component h, s, v and the RGB color model corresponding relation of r, g, b component suc as formula shown in (1)-(3).
Luminance component v does
v=k
max/255
(1)
Saturation degree component s does
(2)
Chromatic component h does
h=h×60;h=h+360(h<0)
(3)
Wherein, and h ∈ [0,360), s ∈ [0,1], v ∈ [0,1], k
Max=max (r, g, b), k
Min=min (r, g, b), δ=k
Max-k
Min
(2) choose black, white, blue, yellow 4 kinds of colors as the benchmark color, according to formula (4) calculate R (x, y) in the distance of each pixel color value and 4 kinds of benchmark colors, be the new color value of this point with the benchmark color of this pixel distance minimum.Table 1 is the rgb value and corresponding HSV value of 4 kinds of benchmark colors.
Distance between two kinds of colors is:
d
1=(v
1-v
2)
2+(s
1×cos(h
1)-s
2×cos(h
2))
2+(s
1×sin(h
1)-s
2×sin(h
2))
2
Wherein, (h
1, s
1, v
1) and (h
2, s
2, v
2) be respectively HSV value before the correction of treating my good adjusting point and as the HSV value of benchmark color, δ is a scale-up factor, δ ∈ (0,1).
The rgb value of table 14 kind of benchmark color and corresponding HSV value
(x, y) (as shown in Figure 5) extract blueness, yellow zone of concentrating as the vehicle candidate region for step 3, the revised image I 2 of scanning.
(1) to image I 2 (x y) lines by line scan, with in each row be the location records of point of vehicle background color (common car plate background color is selected blueness as the car plate background color for blue, yellow in this example) in array S, with the point of delegation, be recorded in the same delegation of S.
(2) analyze the point that is among the S in each row, if the distance of two points, thinks then that these two points are continuous less than s_min, 2 are linked to be a line segment; If greater than s_min, then discontinuous, the first last location records of line segment in array L_cur; Judgement with points all in the delegation after, the line segment that is recorded among the L_cur is screened, if line segment length is greater than l_min's; Then with the location records of line segment in array L, with having many line segments in the delegation, empty L_cur among the L; All row in having analyzed S, preferred s_min gets 40 in the present embodiment, and l_min gets 20.
(3) with the line segment initialization rectangle among the array L, each bar line segment is initialized as a rectangle, and horizontal two limits of rectangle are the row number at line segment place, and rectangle is two limits row that are two somes places of line segment number longitudinally, and initialized rectangle is stored among the array R.
(4) merge neighbouring rectangle among the array R; If vertical bar limit of two rectangles differs less than 3 pixels, the last widthwise edge of the following widthwise edge below rectangle of top rectangle is separated by less than 2 pixels, then merges; Rectangle longitudinal edge invariant position after the merging; The widthwise edge of top is the last widthwise edge of former top rectangle, and following widthwise edge is the following widthwise edge of below rectangle, and two original rectangles are removed.
Step 4, verification vehicle region.Judge the current region size, if the zone is then thought license plate area, otherwise cast out within the scope of setting.Rectangle shown in the R storage is judged, if rectangle length greater than w_min, width is then thought license plate area greater than h_min, the relevant position intercepting goes out car plate in original image at last.W_min=40, h_min=20, the intercepting result is as shown in Figure 6.
Claims (4)
1. a license plate locating method specifically comprises the steps:
(1) to the original vehicle image I (x y) carries out pre-service, the vehicle image I1 after being enhanced (x, y);
(2) revise vehicle image I1 (x, y) in the color value of each point, obtain revised vehicle image I2 (x, y);
(3) (x y), extracts blue and yellow zone of concentrating as license plate candidate area to the revised vehicle image I2 of scanning;
(4) verification license plate area is promptly judged the current region size, if license plate area is then thought in the zone within the scope of setting, accomplishes the car plate location.
2. method according to claim 1 is characterized in that, in the said step (1), carries out pretreated detailed process and is:
(1) with license plate image function I (x y) is expressed as following form:
I(x,y)=R(x,y)L(x,y)
Wherein (x y) is the position coordinates of image mid point, I (x, y) expression original vehicle image, R (x, y) expression reflected light component, L (x, y) expression irradiates light component.
(2) to license plate image function I (x, take the logarithm in expression formula both sides y), and then with the irradiates light component and reflected light component is expressed as and form:
LogI=Log(R.L)=LogR+LogL
(3) (x y) does convolution, obtains the image D behind the corresponding n width of cloth LPF to original image I respectively with Gauss's template of n different scale
i(x, y), wherein, i ∈ [1, n], n is a positive integer:
(4) in log-domain, with original image I (x, y) with every width of cloth LPF after image D
i(x y) subtracts each other, and just can obtain the image that the n panel height strengthens frequently:
R
i(x,y)=LogI(x,y)-LogD
i(x,y)
(5) the n panel height that obtains in (4) is strengthened the image weighted sum frequently, the vehicle image I1 after promptly being enhanced (x, y):
3. method according to claim 1 and 2 is characterized in that, in the said step (2), to vehicle image I1 (x, y) in the detailed process revised of the color value of each point be:
(1) image is transformed into the HSV space from rgb space;
(2) choose black, white, blue, yellow 4 kinds of colors as the benchmark color, calculate I1 (x, y) in the distance of each pixel color value and 4 kinds of benchmark colors, be the new color value of this point with the benchmark color of this pixel distance minimum;
Promptly obtain after said pixel color value upgrades and accomplishes revised vehicle image I2 (x, y).
4. according to the described method of one of claim 1-3, it is characterized in that in the said step (3), extraction license plate candidate area detailed process is:
(1) (x y) lines by line scan, and writes down in each row to be the line segment position of car plate background color color, and the pixel separation of two points of color of the same race is regarded as continuous point less than distance threshold to revised vehicle image I2;
(2) every of analytic record line segment is if line segment length is less than length threshold, then with the deletion from record of this line segment;
(3) with the line segment initialization rectangle of record, each bar line segment is initialized as a rectangle, and horizontal two limits of rectangle are the row number at line segment place, and rectangle is two limits row that are two somes places of line segment number longitudinally;
(4) merge adjacent rectangle, merge into size not in specified scope, deletion then.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110379011.1A CN102521587B (en) | 2011-11-24 | 2011-11-24 | License plate location method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110379011.1A CN102521587B (en) | 2011-11-24 | 2011-11-24 | License plate location method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102521587A true CN102521587A (en) | 2012-06-27 |
CN102521587B CN102521587B (en) | 2014-06-04 |
Family
ID=46292498
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110379011.1A Active CN102521587B (en) | 2011-11-24 | 2011-11-24 | License plate location method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102521587B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408430A (en) * | 2014-12-01 | 2015-03-11 | 广东中星电子有限公司 | License plate positioning method and device |
CN104766049A (en) * | 2015-03-17 | 2015-07-08 | 苏州科达科技股份有限公司 | Method and system for recognizing object colors |
CN107292898A (en) * | 2017-05-04 | 2017-10-24 | 浙江工业大学 | A kind of car plate shadow Detection and minimizing technology based on HSV |
CN108022429A (en) * | 2016-11-04 | 2018-05-11 | 浙江大华技术股份有限公司 | A kind of method and device of vehicle detection |
CN108268871A (en) * | 2018-02-01 | 2018-07-10 | 武汉大学 | A kind of licence plate recognition method end to end and system based on convolutional neural networks |
CN117528045A (en) * | 2024-01-04 | 2024-02-06 | 深圳市云影天光科技有限公司 | Video image processing method and system based on video fog-penetrating anti-reflection technology |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1928892A (en) * | 2006-09-20 | 2007-03-14 | 王枚 | Method and device for license plate location recognition, vehicle-logo location recognition and vehicle type |
CN101334837A (en) * | 2008-07-31 | 2008-12-31 | 重庆大学 | Multi-method integrated license plate image positioning method |
US20100014707A1 (en) * | 2005-06-03 | 2010-01-21 | Takuma Nakamori | Vehicle and road sign recognition device |
-
2011
- 2011-11-24 CN CN201110379011.1A patent/CN102521587B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100014707A1 (en) * | 2005-06-03 | 2010-01-21 | Takuma Nakamori | Vehicle and road sign recognition device |
CN1928892A (en) * | 2006-09-20 | 2007-03-14 | 王枚 | Method and device for license plate location recognition, vehicle-logo location recognition and vehicle type |
CN101334837A (en) * | 2008-07-31 | 2008-12-31 | 重庆大学 | Multi-method integrated license plate image positioning method |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104408430A (en) * | 2014-12-01 | 2015-03-11 | 广东中星电子有限公司 | License plate positioning method and device |
CN104408430B (en) * | 2014-12-01 | 2020-01-07 | 广东中星微电子有限公司 | License plate positioning method and device |
CN104766049A (en) * | 2015-03-17 | 2015-07-08 | 苏州科达科技股份有限公司 | Method and system for recognizing object colors |
CN104766049B (en) * | 2015-03-17 | 2019-03-29 | 苏州科达科技股份有限公司 | A kind of object color recognition methods and system |
CN108022429A (en) * | 2016-11-04 | 2018-05-11 | 浙江大华技术股份有限公司 | A kind of method and device of vehicle detection |
CN108022429B (en) * | 2016-11-04 | 2021-08-27 | 浙江大华技术股份有限公司 | Vehicle detection method and device |
CN107292898A (en) * | 2017-05-04 | 2017-10-24 | 浙江工业大学 | A kind of car plate shadow Detection and minimizing technology based on HSV |
CN108268871A (en) * | 2018-02-01 | 2018-07-10 | 武汉大学 | A kind of licence plate recognition method end to end and system based on convolutional neural networks |
CN117528045A (en) * | 2024-01-04 | 2024-02-06 | 深圳市云影天光科技有限公司 | Video image processing method and system based on video fog-penetrating anti-reflection technology |
CN117528045B (en) * | 2024-01-04 | 2024-03-22 | 深圳市云影天光科技有限公司 | Video image processing method and system based on video fog-penetrating anti-reflection technology |
Also Published As
Publication number | Publication date |
---|---|
CN102521587B (en) | 2014-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102375982B (en) | Multi-character characteristic fused license plate positioning method | |
CN109657632B (en) | Lane line detection and identification method | |
CN110197589B (en) | Deep learning-based red light violation detection method | |
CN110619750B (en) | Intelligent aerial photography identification method and system for illegal parking vehicle | |
US8294794B2 (en) | Shadow removal in an image captured by a vehicle-based camera for clear path detection | |
CN102663760B (en) | Location and segmentation method for windshield area of vehicle in images | |
CN105678285B (en) | A kind of adaptive road birds-eye view transform method and road track detection method | |
CN102521587B (en) | License plate location method | |
CN104036262B (en) | A kind of method and system of LPR car plates screening identification | |
US8345100B2 (en) | Shadow removal in an image captured by a vehicle-based camera using an optimized oriented linear axis | |
CN106651872A (en) | Prewitt operator-based pavement crack recognition method and system | |
US20180182099A1 (en) | Identifying and Excluding Blurred Areas of Images of Stained Tissue To Improve Cancer Scoring | |
CN109886896A (en) | A kind of blue License Plate Segmentation and antidote | |
CN103824081B (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
US8319854B2 (en) | Shadow removal in an image captured by a vehicle based camera using a non-linear illumination-invariant kernel | |
CN103093186A (en) | Systems and methods for improving image recognition | |
US20150286883A1 (en) | Robust windshield detection via landmark localization | |
CN102867185B (en) | Method and system for identifying automobile tire number | |
CN103116751A (en) | Automatic license plate character recognition method | |
CN101334836A (en) | License plate positioning method incorporating color, size and texture characteristic | |
CN103077617A (en) | Pedestrian crosswalk intelligent traffic light monitoring system and method based on computer vision | |
CN103077384A (en) | Method and system for positioning and recognizing vehicle logo | |
CN102509098A (en) | Fisheye image vehicle identification method | |
CN102364496A (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
CN105701483A (en) | Urban boundary extraction method fusing multispectral remote sensing data and night light remote sensing data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |